Cocaine dependence is an entrenched public health problem. Progress has been made in developing efficacious behavioral therapies, but considerable room for improvements remains. The primary focus of this competing continuation is to improve the voucher component of the CRA + Vouchers treatment, one of the most reliably efficacious interventions for cocaine dependence. We also plan to examine the influence of treatment on patient involvement with healthy alternatives to cocaine use and research brief educational interventions to increase patient knowledge about HIV/AIDS and HCV infection. The voucher component is typically offered as a 12-week intervention where patients earn vouchers exchangeable for retail items contingent on cocaine abstinence. Evidence from the current funding period suggests that the voucher program can be reduced to 6 weeks and remain efficacious. Because we use an escalating schedule of voucher payments, reducing program duration from 12 to 6 weeks reduces maximal voucher earnings by 70%. Such cost savings dramatically increase the potential for dissemination to community clinics. They also increase flexibility for parametric changes such as increases in voucher value without resulting in exorbitant overall costs. We propose two randomized clinical trials. In Trial 1, cocaine-dependent outpatients will receive 24 weeks of CRA therapy combined with (a) 6 weeks of abstinence-contingent vouchers, (b) 12 weeks of abstinence-contingent vouchers, or (c) 6 weeks of non-contingent vouchers. Trial 1 will provide a rigorous test of the efficacy of the 6-week voucher intervention. In Trial 2, cocaine-dependent outpatients will receive 24 weeks of CRA therapy combined with either 6 weeks of abstinence-contingent vouchers set at (a) one-third usual value, (b) usual value, or (c) > 3-fold usual value. Trial 2 is designed to examine the influence of voucher value on continuous cocaine abstinence within this 6-week voucher arrangement. Sustained abstinence during treatment is the best predictor of post-treatment abstinence. By restricting the intervention to 6 weeks, we can explore a greater than 3-fold increase in voucher value while keeping the overall maximal costs equal to our usual 12-week voucher program. In both trials we also plan to examine whether voucher-based treatment increases involvement with healthy alternatives to cocaine-produced reinforcement that may influence the likelihood of post-treatment cocaine abstinence and investigate the efficacy of brief educational interventions for increasing HIV/AIDS and HCV knowledge. Overall, the proposed trials have the potential to more effectively target and reduce the cost of incentive-based treatments for cocaine dependence, enhance understanding of how vouchers may produce longer-term cocaine abstinence, and validate methods to educate this at-risk population about HIV/AIDS and HCV infection.